Abstract
Background
Dementia affects 50 million globally. Combining hematologic inflammatory parameters with high-density lipoprotein cholesterol (HDL) has shown potential for predicting neurological diseases, but its link to cognitive function in older individuals is unexplored.
Objective
This study aimed to evaluate the relationships between these novel biomarkers and cognitive impairment in older adults.
Methods
This retrospective study used data of adults aged ≥60 years from the National Health and Nutrition Examination Survey (NHANES) 2011–2014. The analysis examined associations between cognitive impairment and total cholesterol (TC)/HDL, triglyceride (TG)/HDL, monocyte/HDL, neutrophil/HDL, lymphocyte/HDL, and platelet/HDL ratios, using univariate and multivariable logistic regression. Cognitive function was assessed using CERAD, Animal Fluency Test, and Digit Symbol Substitution Test (DSST).
Results
Data from 1379 participants (representing 25,173,874 persons in the US) were analyzed. In the multivariable analysis, compared to the lowest quartile, the highest quartile of neutrophil/HDL ratio was significantly associated with increased odds of low cognitive performance (aOR = 1.85) assessed by DSST. In addition, a unit increase in lymphocyte/HDL was significantly associated with higher odds of low cognitive performance (aOR = 1.16). In stratified analyses, neutrophil/HDL was associated with low cognitive performance in participants without CVD (aOR = 2.09), without CKD (aOR = 2.74), and with or without hypertension (aOR = 3.50 and 4.28). Lymphocyte/HDL was significant only among those without hypertension (aOR = 1.93).
Conclusions
Increased neutrophil/HDL and lymphocyte/HDL ratios were significantly associated with lower cognitive performance in older adults, particularly among those without major comorbidities, suggesting their potential for identifying cognitive risk in healthier populations.
Keywords
Introduction
Approximately 50 million people worldwide are affected by dementia, a number expected to triple by 2050. 1 Alzheimer's disease (AD), characterized by progressive cognitive deterioration and memory loss, stands as the most prevalent form of dementia globally. 2 Within the older population, cognitive impairment not only compromises individual autonomy and quality of life but also poses substantial societal and economic challenges, underscoring the urgency to comprehensively address its underlying determinants. 3 At present, there are no specialized medications for reversing AD, merely postpone its onset and advancement. 4
Numerous studies have documented the relationships between serum lipid profiles and the onset, development, and outcomes of cognitive dysfunction. 4 Studies have shown that high levels of low-density lipoprotein (LDL), total cholesterol (TC), and triglycerides (TG) are negatively correlated with cognitive performance. Conversely, high-density lipoprotein (HDL) displays a unique relationship, is positively associated with cognitive function.5–8 Previous studies in aging women further support HDL's beneficial role in verbal learning and memory performance, suggesting its early impact on healthy brain aging. 6 On the other hand, efforts have been made to integrate blood inflammatory markers with HDL to improve prognostic capabilities in various conditions such as coronary artery disease, diabetes, and malignancies.9–11 For instance, ratios such as monocyte to HDL ratio and neutrophil to HDL ratio not only reflect HDL's anti-inflammatory and antioxidant benefits but also the pro-inflammatory status revealed by blood cells. These biomarkers have demonstrated significant prognostic values, particularly in neurological diseases.12,13 Nonetheless, the relationships between these innovative parameters and cognitive function in older individuals have not been studied yet, creating a compelling area for further research.
Therefore, this study aimed to bridge these gaps by conducting a comprehensive analysis, seeking to elucidate potential correlations between total cholesterol to HDL ratio, alongside an array of ratios of serum inflammatory parameters and HDL, and cognitive impairment in older adults.
Methods
Data source
This study retrospectively analyzed longitudinal data sourced from the United States National Health and Nutrition Examination Survey (NHANES) database. Administered by the National Center for Health Statistics (NCHS), a division of the Centers for Disease Control and Prevention (CDC), this program serves as the primary repository (http://www.cdc.gov/nchs/nhanes). NHANES conducts comprehensive surveys combining interviews and physical examinations to assess the health and nutritional status of both adults and children. Employing a sophisticated, multistage design, NHANES gathers and analyzes data representative of the non-institutionalized US population. Participants are initially interviewed at home and later invited to undergo a thorough examination at a mobile examination center (MEC), encompassing physical assessments, specialized measurements, and laboratory tests. Leveraging NHANES data facilitates robust population-level research on various health-related concerns.
Ethics consideration
The NHANES program underwent rigorous review and approval by the NCHS Research Ethics Review Board. Each participant in the survey provided informed consent by signing a consent form. Researchers are authorized by the NCHS to access and utilize the data, which are made available for research purposes. All NHANES data released by the NCHS are de-identified, ensuring participant anonymity during analysis. Therefore, this study's analysis of the data did not require additional ethical approval or specific informed consent.
Study population selection
Figure 1 illustrates the flow diagram of participant selection. Older participants aged ≥ 60 years in the NHANES 2011–2014 (two cycles), with complete data on components of the biomarkers of interest, and with data from cognitive tests were eligible for inclusion. These two study cycles were specifically selected because they are the sole cycles within NHANES that evaluated cognitive function. Subjects with missing data on sex or sample weights were excluded.
Flow diagram of study population selection.
Cognitive function assessment
In 2011–2014, NHANES conducted a series of assessments of cognitive function in participants aged 60 years and older. The questionnaire was administered by trained interviewers at the end of the face-to-face private interview in the Mobile Examination Center. These cognitive functioning tests contained: The Consortium to Establish a Registry for Alzheimer's Disease (CERAD), the Animal Fluency Test, as well as the Digit Symbol Substitution Test (DSST).
Individuals scoring at or below the 25th percentile on each cognitive test within the total study population were categorized as having low cognitive performance.
HDL and hematologic inflammation markers
Our analysis examined the potential relationships between TC/HDL, neutrophil/HDL, lymphocyte/HDL, platelet/HDL, monocyte/HDL ratios, and the abovementioned cognitive function tests. The NHANES CBC Profile gathered the hematologic parameters using Beckman Coulter's method for counting and sizing, along with an automated diluting and mixing device for sample processing. Hemoglobin levels were measured using a single-beam photometer. The WBC differential utilized VCS technology. TC and HDL levels were also assessed using blood samples obtained during the study visit. Each factor was categorized into four quartiles, with quartile 1 closest to the normal condition.
Covariates
Information regarding age, sex, race, income-to-poverty ratio, marital status, and education level was meticulously gathered through in-person interviews conducted by skilled interviewers.
The Body Mass Index (BMI) values were derived from NHANES examination measurements, calculated as the ratio of body weight in kilograms to the square of height in meters. These values were categorized as <25.0, 25 to 29.9, and ≥30 kg/m2. Body weight was measured using an electronic load cell scale while standing height was measured with a fixed stadiometer.
Smoking status was categorized into a non-smoker, former smoker, or current smoker based on specific criteria: individuals who had smoked fewer than 100 cigarettes in their lifetime were classified as non-smokers, those who had smoked >100 cigarettes but were not currently smoking were termed former smokers, and those who had smoked >100 cigarettes and responded “yes” to the question “Do you smoke now?” were considered current smokers.
Participants with diabetes mellitus (DM) were identified using several criteria, including a ‘yes’ response to questions about a previous diagnosis or the use of diabetes medications, or by meeting specific laboratory parameters such as HbA1c ≥ 6.5%, fasting glucose ≥ 126 mg/dL, or random glucose levels ≥ 200 mg/dL.
Systolic (SBP) and diastolic blood pressure (DBP) were measured thrice following standardized protocols and then averaged for analysis. Hypertension was defined by affirmative responses to questions regarding diagnosis or medication intake, or by averaging three consecutive measurements showing SBP ≥ 140 mmHg or DBP ≥ 90 mmHg.
CVD history was defined as the presence of any self-reported diagnosis of the following conditions: coronary heart disease, angina pectoris, heart attack, stroke, or congestive heart failure. This information was collected based on the health-related condition questions asked in the survey.
Chronic respiratory disease was self-reported based on participants’ acknowledgment of a doctor or health professional diagnosing them with asthma, chronic bronchitis, or emphysema.
Glomerular filtration rate (GFR) was estimated using the 4-variable Modification of Diet in Renal Disease (MDRD) Study equation, specifically the IDMS-traceable MDRD Study equation that incorporates standardized creatinine levels. Chronic kidney disease (CKD) was defined as having an estimated GFR <60 ml/min/1.73 m2.
Statistical analysis
NHANES uses a complex, multistage, probability sampling design to assure national representation, wherein sampling weights (WTINT2YR), pseudo-stratum (SDMVSTRA), and pseudo-cluster (SDMVPSU) provided by NHANES were applied in all analyses as guided by the NCHS. The weighted mean and standard error were presented for continuous variables; unweighted number and weighted proportion were presented for categorical variables. All statistical analyses were performed using SAS statistical software (version 9.4, SAS Inc., Cary, NC, USA). p-value for group comparisons was calculated by using the PROC SURVEYFREQ and SURVEYREG on categorical and continuous data.
The outcome variable, low cognitive performance, is binary for each cognitive test. We utilized Binary Logistic Regression with PROC SURVEYLOGISTIC to assess the associations between the biomarkers and low cognitive performance. Multivariable model 1 adjusted for age (continuous) and sex; and model 2 adjusted for age (continuous), sex, and education level. A two-sided P-value of <0.05 was regarded as statistically significant.
Given the significant associations observed in the overall population, we subsequently conducted subgroup analyses stratified by the presence or absence of specific comorbid conditions (i.e., CVD, CKD, and hypertension) to examine the associations between two biomarkers—neutrophil/HDL and lymphocyte/HDL ratios—and low cognitive function, as assessed by the DSST. In addition to age- and sex-adjusted models, we applied fully adjusted multivariable models that included all variables found to be significantly different in univariate analyses, excluding those with high collinearity.
Results
Study population
Figure 1 demonstrates the flow diagram of the study participants’ selection. Data from a total of 19,931 participants were extracted. A total of 1381 participants aged ≥ 60 years with complete data on the three cognitive tests (i.e., CERAD, Animal Fluency Test, and DSST) along with the biomarkers of interest (i.e., the lipids and the hematologic indices) were included. Two participants with abnormal data on the biomarkers of interest were excluded. Consequently, 1379 participants were included in the analysis, representing 25,173,874 US older adults living in the community (Figure 1).
Characteristics of the study population
Table 1 summarizes the characteristics of the study population by whether or not having cognitive impairment. The mean age of the participants was 69.2 ± 0.3 years, 53.6% were females and the majority were non-Hispanic Whites (81.3%). The most prevalent comorbidity was hypertension (64.0%). A total of 394 individuals had cognitive impairment assessed by CERAD, 406 by Animal Fluency Test, and 371 by DSST. Higher neutrophil/HDL had a significantly higher risk of low cognitive impairment test (Table 1).
Characteristics of the study population by status of cognitive performance.
CERAD: Consortium to Establish a Registry for Alzheimer's disease; CVD: cardiovascular disease; DSST: Digit Symbol Score Test; TC: total cholesterol; TG: triglyceride; HDL: high-density lipoprotein cholesterol; HbA1c: glycated hemoglobin.
Continuous variables are presented as mean ± SE; categorical variables are presented as unweighted counts (weighted percentage).
p-values < 0.05 are shown in bold.
Associations between the biomarkers and low cognitive performance
Supplemental Table 1 demonstrates the associations between the biomarkers and low cognitive performance.
When assessed by CERAD, elevated neutrophil/HDL ratio was significantly associated with an increased odd of low cognitive performance (adjusted odd ratio [aOR] = 1.14, 95% confidence interval [CI]: 1.10–1.29) after adjusting for age and sex in model 1. Higher quartile of lymphocyte/HDL ratio were significantly associated with increased odds of low cognitive performance (quartile 3: aOR =1.45, 95%CI: 1.00–2.91; quartile 4: aOR = 1.73, 95%CI: 1.07–2.77) as compared to the lowest quartile. However, after further adjusting for education level in model 2, the significances were gone.
When assessed by the Animal Fluency Test, the results revealed that neutrophil/HDL quartile 4 was significantly associated with increased odds of low cognitive performance (aOR = 1.72, 95%CI: 1.21–2.45) after adjusting age and sex in model 1. After further adjusting for education level in model 2, the significance was gone.
When assessed by DSST, elevated neutrophil/HDL (aOR = 1.28, 95% CI: 1.08–1.51), monocyte/HDL (aOR = 1.20, 95% CI: 1.00–1.44), and lymphocyte/HDL (aOR = 1.31, 95% CI: 1.09–1.59) were significantly associated with increased odds of low cognitive performance in age- and sex-adjusted model (model 1). After further adjusting for education level in model 2, the results showed that neutrophil/HDL quartile 4 was significantly associated with increased odds of low cognitive performance (aOR = 1.85, 95% CI: 1.03–3.32), and elevated lymphocyte/HDL was significantly associated with higher odds of low cognitive performance (aOR = 1.16, 95% CI: 1.01–1.34) (Supplemental Table 1).
Associations between TC, TG, neutrophils, monocytes, platelets, lymphocytes (not in ratios to HDL) and low cognitive performance were examined. In model 2, no significant associations with low cognitive performance were observed, except for TC with the Animal Fluency Test or DSST (Supplemental Table 2).
Given that neutrophil/HDL and lymphocyte/HDL ratios were significantly associated with a higher likelihood of low cognitive performance, we conducted a receiver operating characteristic (ROC) curve analysis to evaluate their predictive accuracy for low cognitive performance, as measured by the DSST. The results show AUC values of 0.576 for neutrophil/HDL and 0.597 for lymphocyte/HDL (Supplemental Table 3).
Associations between neutrophil/HDL, lymphocyte/HDL, and low cognitive performance assessed by DSST, stratified by CVD, CKD, and hypertension
We further carried out a stratified analysis by CVD history, CKD, and hypertension. The results are shown in Tables 2–4.
Association between neutrophil/HDL, lymphocyte/HDL, and low cognitive performance by DSST, stratified by CVD history.
CVD: cardiovascular disease; DSST: Digit Symbol Score Test; HDL: high-density lipoprotein cholesterol; aOR: adjusted odds ratio; CI: confidence interval; Ref: reference.
p-values < 0.05 are shown in bold.
Z-score was standardized by subtracting the mean and then dividing the result by the standard deviation.
Model A: Adjusted for age (continuous) and sex.
Model B: Adjusted for related variables of p < 0.05 in univariate analysis (excluding stratification factors and variables with strong correlations), including age (continuous), race, poverty income ratio, and excessive alcohol consumption.
Association between neutrophil/HDL, lymphocyte/HDL, and low cognitive performance by DSST, stratified by CKD status.
CKD: chronic kidney disease; DSST: Digit Symbol Score Test; HDL: high-density lipoprotein cholesterol; aOR: adjusted odds ratio; CI: confidence interval; Ref: reference.
p-values < 0.05 are shown in bold.
Z-score was standardized by subtracting the mean and then dividing the result by the standard deviation.
Model A: Adjusted for age (continuous) and sex.
Model B: Adjusted for related variables of p < 0.05 in univariate analysis (excluding stratification factors and variables with strong correlations), including age (continuous), race, poverty income ratio, and excessive alcohol consumption.
Association between neutrophil/HDL, lymphocyte/HDL, and low cognitive performance by DSST, stratified by hypertension status.
DSST: Digit Symbol Score Test; HDL: high-density lipoprotein cholesterol; aOR: adjusted odds ratio; CI: confidence interval; Ref: reference.
p-values < 0.05 are shown in bold.
Z-score was standardized by subtracting the mean and then dividing the result by the standard deviation.
Model A: Adjusted for age (continuous) and sex.
Model B: Adjusted for related variables of p < 0.05 in univariate analysis (excluding stratification factors and variables with strong correlations), including age (continuous), race, poverty income ratio, and excessive alcohol consumption.
CVD
After adjustment by age and sex (model A), the highest quartile of neutrophil/HDL ratio was significantly associated with low cognitive function only in individuals without CVD history (aOR = 2.46, 95% CI: 1.37–4.40), with no significant association observed in those with CVD history. The highest quartile of lymphocyte/HDL was significantly associated with low cognitive function among individuals without CVD history (aOR = 2.44, 95% CI: 1.35–4.41) and with CVD history (aOR = 2.49, 95% CI: 1.06–5.82).
In model B, which included covariates that were statistically significant in univariate analyses, only the association between the highest quartile of neutrophil/HDL and low cognitive function in individuals without a history of CVD remained statistically significant (aOR = 2.09, 95% CI: 1.01–4.31) (Table 2).
CKD
After adjusted by age and sex (model A), the third quartile (aOR = 1.71, 95% CI: 1.02–2.86) and the highest quartile (aOR = 2.51, 95% CI: 1.48–4.24) of neutrophil/HDL were significantly associated with low cognitive function among individuals without CKD. Similarly, the highest quartile (aOR = 2.53, 95% CI: 1.55–4.12) of lymphocyte/HDL ratio was significantly associated with low cognitive function among individuals without CKD. However, no significant associations were observed for either biomarker among individuals with CKD.
In model B, which included covariates that were statistically significant in univariate analyses, only the highest quartile (aOR = 2.74, 95% CI: 1.54–4.9) of neutrophil/HDL remained significantly associated with low cognitive function among individuals without CKD (Table 3).
Hypertension
After adjusted for age and sex in Model A, the highest quartile of neutrophil/HDL was significantly associated with low cognitive function among individuals without hypertension (aOR = 3.90, 95% CI: 1.19–12.83) and with hypertension (aOR = 1.84, 95% CI: 1.00–3.38). Similarly, the highest quartile of lymphocyte/HDL ratio was significantly associated with low cognitive function among individuals without hypertension (aOR = 3.40, 95% CI: 1.26–9.21) and with hypertension (aOR = 2.00, 95% CI: 1.16–3.45).
In Model B, which included covariates that were statistically significant in univariate analyses, the highest quartile of neutrophil/HDL remained significantly associated with low cognitive function both among individuals without hypertension (aOR = 4.28, 95% CI: 1.13–16.20) and those with hypertension (aOR = 3.50, 95% CI: 1.15–10.63). Additionally, the highest quartile of lymphocyte/HDL remained significantly associated with low cognitive function in participants without hypertension (aOR = 1.93, 95% CI: 1.05–3.55) but not among those with hypertension (Table 4).
Discussion
This study investigated the potential relationships between cognitive impairment and several novel biomarkers integrating both HDL and hematological inflammatory parameters among US older adults dwelling in the community. The analytic results in for overall study population revealed that high neutrophil/HDL and lymphocyte/HDL are associated with low cognitive performance. Notably, evaluations of neutrophil and lymphocyte counts alone did not show a significant association with low cognitive function, underscoring the rationale for using their ratios to HDL in this context. Further stratified analyses based on the fully adjusted model revealed that neutrophil/HDL remained significantly associated with low cognitive function among individuals without CVD and without CKD, but not among those with these conditions. Regarding hypertension, neutrophil/HDL was significantly associated with low cognitive function in both hypertensive and non-hypertensive individuals. Among all subgroups, lymphocyte/HDL only showed a significant association with low cognitive function among those without hypertension. These findings suggest that neutrophil/HDL may serve as a more broadly applicable marker, and the associations between these biomarkers and cognitive impairment are more prominent in relatively healthier individuals. This indicates their potential utility for early detection of cognitive decline before the onset of significant comorbidities.
Our findings contribute to the growing body of literature exploring the relationship between readily obtainable hematologic indices and cognitive function,14,15 advancing understanding of potential mechanisms underlying AD. Earlier research, such as that by Kivipelto et al. (2006), 16 has established a link between midlife hypercholesterolemia and increased AD risk, suggesting that cholesterol may influence amyloid-β (Aβ) deposition, oxidative stress, and neuroinflammation, all key processes in AD pathology. However, the precise role of cholesterol in AD remains complex and sometimes contradictory. Amongst, HDL, known for its antioxidative properties, 17 has been less extensively studied and has produced inconsistent findings.6,18,19 On the other hand, a study by Kuyumcu et al. (2012) 20 identified elevated neutrophil-lymphocyte ratios (NLR) as markers of inflammation in AD, reinforcing the link between systemic inflammation and cognitive decline. By integrating inflammatory markers with HDL, our study expands on previous work by demonstrating that higher neutrophil/HDL and lymphocyte/HDL ratios are associated with lower cognitive performance in older adults. These findings suggest that elevated ratios may reflect both reduced antioxidant capacity and heightened chronic inflammation (as indicated by higher neutrophil/HDL), potentially accelerating cognitive decline. This mechanism may underlie the observed association.
AD has distinctive pathological features in the accumulation of Aβ plaques and the formation of neurofibrillary tangles. 21 In the context of individuals exhibiting normal cognitive function, microglia, which are the resident brain immune cells, migrate towards Aβ plaques to engage in clearing Aβ.22,23 However, this initial defense mechanism can become compromised over extended periods, leading to microglial hypertrophy and a subsequent decline in their efficacy in Aβ phagocytosis. 24 Consequently, peripheral leukocytes are recruited to the sites of Aβ plaque deposition to eliminate Aβ, and this recruitment process is intricately regulated by various cytokines. While these cytokines play a pivotal role in facilitating Aβ clearance by recruiting peripheral leukocytes, they also contribute to the deleterious effects of persistent inflammation, thereby influencing the progression of AD pathology.25,26 Clinically, immune dysregulation was observed in AD patients, including higher neutrophil levels, an increased neutrophil-to-lymphocyte ratio, and a decreased lymphocyte count.20,27,28
Beyond the realm of its antioxidative properties, HDL has been found to exert regulatory effects on neutrophil activation, ultimately reducing neutrophil proliferation and migration.29,30 As previously mentioned, studies examining the relationship between HDL and cognition have produced inconsistent results. For instance, a cohort study found that among initially healthy older adults, those with high serum HDL levels at recruitment had a 27% higher risk of developing dementia during follow-up. 31 Given the complex etiology of AD, which involves altered lipid metabolism, inflammation, oxidative damage, and other pathological processes, combining the inflammation index with HDL levels, as we have done in the current analyses, may provide a more comprehensive approach for predicting risks. Further, the neutrophil/HDL ratio was shown to be a useful indicator for cardiovascular disease, 32 coronary stenosis, 33 an acute coronary syndrome in patients with type 2 diabetes, 10 and many other diseases. The same goes for the lymphocyte/HDL ratio, which is associated with cardiometabolic diseases 34 and mortality in sepsis patients. 35 Although both indexes can be applied in many conditions, little is known in the field of dementia. Here, we provide evidence that higher neutrophil/HDL and lymphocyte/HDL ratios are associated with lower cognitive performance among older adults. Although this study did not directly assess AD, the observed associations between these markers and cognitive performance may have implications for early detection of neurodegenerative processes.
Three distinct sets of data related to low cognitive performance were extracted from the NHANES. The CERAD was employed to evaluate both immediate and delayed learning capacities concerning new verbal information. 36 The Animal Fluency test, focusing on categorical verbal fluency, served as a measure of executive function. 37 Additionally, the DSST gauged information processing speed, sustained attention, and working memory. 38 These assessments target diverse cognitive domains, all of which are related to the consequences of AD. The findings of our investigation revealed a notable association between the DSST results and studied biomarkers. This observation implies a potential specificity of impact in these cognitive domains. The significant results observed for the DSST, but not for the CERAD or Animal Fluency Test, may be due to the different cognitive domains each test measures. The DSST primarily assesses processing speed, sustained attention, and working memory, which might be more sensitive to systemic inflammation, 39 explaining its significant associations with the inflammatory ratios (neutrophil/HDL and lymphocyte/HDL). In contrast, CERAD evaluates memory (immediate and delayed recall), and Animal Fluency tests executive function and verbal fluency, which may not be as directly affected by inflammation or lipid metabolism in early cognitive decline. These cognitive domains may be influenced by other factors or require more time to show associations, highlighting the specificity of the DSST in detecting early inflammatory effects on cognitive function. Nevertheless, further in-depth studies are still warranted to elucidate the intricate details of this noteworthy observation.
Additionally, our stratified analyses revealed that the associations between elevated neutrophil/HDL ratios and low cognitive function were more pronounced among individuals without CVD, CKD, or hypertension. Similarly, the association between lymphocyte/HDL and cognitive impairment appeared stronger in those without hypertension. These findings suggest that the value of these biomarkers may be more evident in relatively healthier individuals. One possible explanation is that chronic conditions such as CVD and CKD are already associated with elevated systemic inflammation and oxidative stress, which may raise the baseline levels of inflammatory markers and reduce the relative discriminatory power of neutrophil/HDL or lymphocyte/HDL ratios.40,41 Moreover, these chronic diseases may involve complex physiological adaptations, medication effects, or compensatory mechanisms that could obscure or modify the direct relationship between inflammation and cognitive function. Another consideration is that in populations with advanced comorbidities, cognitive impairment may result from multifactorial or disease-specific pathways (e.g., vascular lesions, uremic toxins) that are less directly tied to immune-inflammatory profiles captured by these ratios. In contrast, among healthier individuals, subclinical inflammation reflected by altered leukocyte-to-HDL ratios might play a more detectable role in early cognitive decline. Nonetheless, these interpretations remain speculative and warrant further investigation through mechanistic studies or longitudinal analyses to better understand the interplay between systemic inflammation, comorbid conditions, and neurocognitive health.
Strength and limitations
This study stands out due to its effective utilization of NHANES data, which represents a wide-ranging and diverse spectrum of individuals across the US community. This broad scope enhances the findings’ robustness and enables a high degree of generalizability to the entire US population. Nonetheless, this study bears several limitations. Firstly, its reliance solely on data from the US population restricts direct generalization to other demographic groups, limiting the broader applicability of its findings. Additionally, the use of single measurements for the hematologic and lipid biomarkers rather than repeated assessments could potentially introduce bias due to the possible variations over time, thereby impacting the accuracy of the analysis. Lastly, although we found significant associations between certain biomarkers and low cognitive function after adjusting for age, sex and education level, the ROC analysis assessing the predictive accuracy of neutrophil/HDL and lymphocyte/HDL alone (without additional covariates) indicated limited accuracy. This suggests that these biomarkers alone may not be sufficient for reliable prediction, underscoring the need for a more comprehensive model that includes other relevant factors to improve predictive accuracy.
Conclusion
The study reveals significant associations between increased neutrophil/HDL and lymphocyte/HDL, and lower cognitive performance in older US adults. These associations appeared more pronounced in individuals without major comorbidities, suggesting that these biomarkers may be particularly useful for identifying cognitive risk in relatively healthy populations. This suggests that, pending further validation through longitudinal studies, monitoring these ratios could potentially serve as a biomarker for cognitive health in the older population.
Footnotes
Acknowledgments
The authors have no acknowledgments to report.
Author contributions
Ethical considerations
The NHANES program underwent rigorous review and approval by the NCHS Research Ethics Review Board. Researchers are authorized by the NCHS to access and utilize the data, which are made available for research purposes. All NHANES data released by the NCHS are de-identified, ensuring participant anonymity during analysis. Therefore, this study's analysis of the data did not require additional ethical approval or specific informed consent.
Consent to participate
Each participant in the survey provided informed consent by signing a consent form.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
All data analyzed during this study are included in this published article
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
